Multiscale Wavelet and Graph Network With Spectral Self-Attention for Hyperspectral Image Classification
Hyperspectral image (HSI) classification has gained increasing attention in remote sensing due to its finegrained spectral information. However, existing methods still face significant challenges in preserving high-frequency details, modeling long-range dependencies, and integrating spectral, spatia...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/11018235/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849417625944195072 |
|---|---|
| author | Anyembe C. Shibwabo Zou Bin Tahir Arshad Jorge Abraham Rios Suarez |
| author_facet | Anyembe C. Shibwabo Zou Bin Tahir Arshad Jorge Abraham Rios Suarez |
| author_sort | Anyembe C. Shibwabo |
| collection | DOAJ |
| description | Hyperspectral image (HSI) classification has gained increasing attention in remote sensing due to its finegrained spectral information. However, existing methods still face significant challenges in preserving high-frequency details, modeling long-range dependencies, and integrating spectral, spatial, and frequency-domain features. In this work, we propose MWGN-SSA, a powerful network designed to enhance HSI classification by fusing multidomain features. MWGN-SSA consists of three core modules: a multiscale learnable wavelet network (MLWN), a window-based spectral self-attention (WSSA) mechanism, and a deep-hop graph convolutional network (DH-GCN). First, MLWN adaptively decomposes HSIs into frequency subbands, retaining critical high-frequency textures for small or spectrally subtle targets. Second, WSSA captures both local and global spectral correlations using a windowed self-attention scheme. Third, DH-GCN constructs a deep graph structure to model spatial topology and overcome oversmoothing. A feature integration module combines outputs from all branches for final prediction. Extensive experiments on four benchmark datasets demonstrate that MWGN-SSA achieves superior accuracy and robustness, particularly in complex and imbalanced HSI scenes. |
| format | Article |
| id | doaj-art-e3d64d7d47b34734b30ccedee3d2e77a |
| institution | Kabale University |
| issn | 1939-1404 2151-1535 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj-art-e3d64d7d47b34734b30ccedee3d2e77a2025-08-20T03:32:42ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-0118151161513610.1109/JSTARS.2025.357520711018235Multiscale Wavelet and Graph Network With Spectral Self-Attention for Hyperspectral Image ClassificationAnyembe C. Shibwabo0https://orcid.org/0009-0002-9106-5514Zou Bin1https://orcid.org/0000-0001-6135-3174Tahir Arshad2https://orcid.org/0009-0009-9038-3722Jorge Abraham Rios Suarez3https://orcid.org/0009-0009-2060-119XSchool of Electronics and Information Engineering, Harbin Institute of Technology, Harbin, ChinaSchool of Electronics and Information Engineering, Harbin Institute of Technology, Harbin, ChinaSchool of Electronics and Information Engineering, Harbin Institute of Technology, Harbin, ChinaSchool of Electronics and Information Engineering, Harbin Institute of Technology, Harbin, ChinaHyperspectral image (HSI) classification has gained increasing attention in remote sensing due to its finegrained spectral information. However, existing methods still face significant challenges in preserving high-frequency details, modeling long-range dependencies, and integrating spectral, spatial, and frequency-domain features. In this work, we propose MWGN-SSA, a powerful network designed to enhance HSI classification by fusing multidomain features. MWGN-SSA consists of three core modules: a multiscale learnable wavelet network (MLWN), a window-based spectral self-attention (WSSA) mechanism, and a deep-hop graph convolutional network (DH-GCN). First, MLWN adaptively decomposes HSIs into frequency subbands, retaining critical high-frequency textures for small or spectrally subtle targets. Second, WSSA captures both local and global spectral correlations using a windowed self-attention scheme. Third, DH-GCN constructs a deep graph structure to model spatial topology and overcome oversmoothing. A feature integration module combines outputs from all branches for final prediction. Extensive experiments on four benchmark datasets demonstrate that MWGN-SSA achieves superior accuracy and robustness, particularly in complex and imbalanced HSI scenes.https://ieeexplore.ieee.org/document/11018235/Attention mechanismconvolutional neural network (CNN)feature integrationgraph convolution network (GCN)hyperspectral image (HSI) classification |
| spellingShingle | Anyembe C. Shibwabo Zou Bin Tahir Arshad Jorge Abraham Rios Suarez Multiscale Wavelet and Graph Network With Spectral Self-Attention for Hyperspectral Image Classification IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Attention mechanism convolutional neural network (CNN) feature integration graph convolution network (GCN) hyperspectral image (HSI) classification |
| title | Multiscale Wavelet and Graph Network With Spectral Self-Attention for Hyperspectral Image Classification |
| title_full | Multiscale Wavelet and Graph Network With Spectral Self-Attention for Hyperspectral Image Classification |
| title_fullStr | Multiscale Wavelet and Graph Network With Spectral Self-Attention for Hyperspectral Image Classification |
| title_full_unstemmed | Multiscale Wavelet and Graph Network With Spectral Self-Attention for Hyperspectral Image Classification |
| title_short | Multiscale Wavelet and Graph Network With Spectral Self-Attention for Hyperspectral Image Classification |
| title_sort | multiscale wavelet and graph network with spectral self attention for hyperspectral image classification |
| topic | Attention mechanism convolutional neural network (CNN) feature integration graph convolution network (GCN) hyperspectral image (HSI) classification |
| url | https://ieeexplore.ieee.org/document/11018235/ |
| work_keys_str_mv | AT anyembecshibwabo multiscalewaveletandgraphnetworkwithspectralselfattentionforhyperspectralimageclassification AT zoubin multiscalewaveletandgraphnetworkwithspectralselfattentionforhyperspectralimageclassification AT tahirarshad multiscalewaveletandgraphnetworkwithspectralselfattentionforhyperspectralimageclassification AT jorgeabrahamriossuarez multiscalewaveletandgraphnetworkwithspectralselfattentionforhyperspectralimageclassification |